/*******************************************************************************

                           REPLICATION PACKAGE

THE SPILLOVER (AND DIRECT) EFFECTS OF WORKPLACE CONTACT ON INTERGROUP ATTITUDES

                            ROBERT JENSEN
                    REVIEW OF ECONOMICS AND STATISTICS

                           AUGUST 21, 2025
					   
*******************************************************************************/


Executed using STATA 15.1

The data file needed to reproduce the results are:
	Do File: 	Contact_Replication.do
	Data File:	RESTAT_Replication_Data.dta 	(Stata 15 format)


To replicate, just replace the part in quotes below on line 36 to the directory 
where you saved the files:
	global CONTACT_REPLIC " < your directory here > " ;


/******************************************************************************
		VARIABLE LIBRARY
*******************************************************************************
						
Variable Name			Description
--------------------------------------------------------------------------------
worker_id			ID of the worker who applied for a BPO job. For workers, this
					is their own ID. For family/friends, it is the ID of the
					worker who designated them as a friend/family member.
worker 				=1 if respondent is the applicant/worker, 
					=0 if respondent is family/friend of the applicant/worker.
muslim_partner 			Binary variable, =1 if randomly assigned Muslim partner, 
					=0 if assigned a Hindu partner
age 				Respondent's age, in years
female 				Binary, =1 if respondent is female, =0 otherwise.
fathers_ed 			Respondent's father's highest level of education
mothers_ed 			Respondent's mother's highest level of education
wealth_index 			Index of respondent's household wealth from 1 to 5.
family_size 			# people in respondent's household 					
interact_with_worker 		How often family/friend respondent reports interacting w/ worker: 
					1=Never
					2=Rarely/not very often;
					3=Sometimes
					4=Often
					5=All the time/every day
have_muslim_friend 		Binary, =1 if respondent says they have a Muslim friend
attriter 			Binary, =1 if respondent could not be surveyed at follow-up
how_much_interact 		How often respondent reports interacting with Muslims: 
					1=Never
					2=Rarely/not very often;
					3=Sometimes
					4=Often
					5=All the time/every day
can_be_friends 			Respondent agrees/disagrees with the statement 
				"Hindus & Muslims can be friends": 
					1=Strongly disagree
					2=Somewhat disagree
					3=Neither agree nor disagree
					4=Somewhat agree
					5=Strongly agree
can_trust 			Respondent agrees/disagrees with the statement 
				"Hindus & Muslims can trust each other"
					1=Strongly disagree
					2=Somewhat disagree
					3=Neither agree nor disagree
					4=Somewhat agree
					5=Strongly agree
treated_unfair 			Respondent agrees/disagrees with the statement 
				"Muslims sometimes face discrimination or unfair 
				treatment in society" 
					1=Strongly disagree
					2=Somewhat disagree
					3=Neither agree nor disagree
					4=Somewhat agree
					5=Strongly agree												
quiz 				# Correct answers to questions about Islam (out of 3)

*******************************************************************************/